New to AI? Start here.

Learn the handful of words every AI conversation uses.

1. Artificial intelligenceAn AI system infers from received inputs how to generate predictions, content, recommendations or decisions for explicit or implicit objectives.5 min
2. AI modelAn AI model is the part of an AI system that has learned from data. It is a structure plus a set of learned numbers that turns an input into a prediction or new content.4 min
3. AlgorithmAn algorithm is a precise, step-by-step procedure for getting a result. In AI, learning algorithms are the procedures that turn data into a trained model.4 min
4. Training dataTraining data is the set of examples a machine learning model studies to learn its patterns, kept apart from the examples used to test it.5 min
5. InferenceInference is using a trained AI model on new input to get an answer, such as a label, a number or a written reply, without changing what the model learned.5 min

See how a computer learns from examples instead of rules.

1. Machine learningMachine learning trains software on data so it can find patterns and make useful predictions or generate content for new inputs.5 min
2. FeaturesFeatures are the input facts a machine learning model reads about each example, such as a car's mileage or colour, turned into a list of numbers.5 min
3. VectorsA vector is an ordered list of numbers that you can also picture as an arrow in space. Machine learning stores examples, words and images this way.4 min
4. Feature engineeringFeature engineering turns raw data, like prices and colour names, into the lists of numbers a model can learn from.4 min
5. TrainingTraining is the repeated loop that nudges a model's internal numbers so its predictions get closer to the right answers in its examples.5 min
6. Neural networksA neural network is a model built from layers of simple units that each weigh their inputs, add them up and bend the result, so that together they can learn curved, complex patterns.4 min
7. Deep learningDeep learning is machine learning with neural networks that stack many layers, so each layer can build a more abstract picture of the data than the one before.4 min

Understand what happens between your question and the answer.

1. Generative AIGenerative AI learns patterns from existing data, then uses those patterns to create new text, images, audio, video, code or other content.5 min
2. LLMAn LLM predicts a token or sequence of tokens, sometimes many paragraphs long.3 min
3. TokensTokens are chunks of text processed by text-generation and embedding models.3 min
4. Next-token predictionNext-token prediction estimates which token should follow the tokens already present in a sequence.3 min
5. TemperatureTemperature enters softmax by dividing each logit by T.3 min
6. Context windowA context window is all the text a language model can reference while generating a response, including the response itself.3 min
7. HallucinationA hallucination is generated content that sounds plausible but is false, unsupported by evidence, or inconsistent with the given input.3 min

See why a chatbot follows instructions instead of just finishing your sentence.

1. PretrainingIn one transfer-learning setup, a model is pretrained on a data-rich task before fine-tuning on a downstream task.3 min
2. Instruction tuningInstruction tuning fine-tunes a pretrained model on many tasks written as instructions paired with desired responses.3 min
3. RLHFRLHF uses human comparisons to learn a reward signal, then optimizes a model to produce responses that score better under that signal.4 min

Tell frontier, open-weight, closed and small models apart, and what their sizes mean.

1. ParametersParameters are the model values that training can change to improve how inputs map to outputs.3 min
2. Mixture of expertsA mixture-of-experts layer uses a learned gate to select a sparse combination of expert subnetworks for each input.3 min
3. Foundation modelsA foundation model is trained on broad data and can be adapted for many downstream tasks.3 min
4. Frontier modelsFrontier models are highly capable general-purpose AI models at or beyond the capabilities of the most advanced current models.3 min
5. Open-weight modelsAn open-weight model makes its trained weights publicly available for download.3 min
6. Closed modelsA fully closed model keeps its weights and code proprietary for internal use.3 min
7. Small language modelsSmall-model research includes sub-billion-parameter language models for mobile deployment.3 min

See why AI runs on GPUs and other special chips, and what compute means.

1. CPUsA CPU is the chip that runs a computer's software, carrying out instructions with just a few cores backed by lots of cache memory.5 min
2. GPUsA GPU is a chip built to run many similar calculations at the same time, which suits the matrix maths inside neural networks.5 min
3. TPUsTPUs are chips Google designed to speed up the maths of machine learning, rented out through Google Cloud.5 min
4. NPUsAn NPU is a part of a chip built to speed up AI models at low power.5 min
5. AI acceleratorsAI accelerators are chips built to speed up the maths inside AI models, which is largely multiply-add sums.5 min
6. AWS TrainiumAWS Trainium is a computer chip Amazon designed for training and running AI models, rented through special Amazon EC2 cloud servers.4 min
7. AWS InferentiaAWS Inferentia is a family of computer chips that Amazon designed to run trained AI models cheaply and quickly in its cloud.4 min
8. FLOPsFLOPs count the small arithmetic steps, like one multiply or one add on decimal numbers, that a computer does to train an AI model.5 min

See how AI reads photos and speech, talks back, and makes pictures, video and songs.

1. Multimodal modelsA multimodal model can process more than one type of input, such as text, images, audio or video.4 min
2. Vision-language modelsA vision-language model can take visual data and text as input and produce text as output.4 min
3. Speech-to-text modelsA speech-to-text model can transcribe a speech utterance into written characters.3 min
4. Text-to-speech modelsA text-to-speech model synthesizes speech directly from text.3 min
5. Voice agentsA voice agent is an AI app you talk to out loud; it listens, works out what you want, can use tools, and answers in speech.5 min
6. Text-to-image modelsA text-to-image model can use text as a conditioning input to an image generator.3 min
7. Diffusion modelsA diffusion model learns to reverse a process that gradually adds noise to data.3 min
8. Image editing modelsAn image editing model changes a picture you already have.4 min
9. Text-to-video modelsGiven a text prompt, a text-to-video model generates a video.3 min
10. Music generation modelsA music generation model can generate music in the raw-audio domain.3 min

Get better answers, check them, and know what not to share.

1. Prompt engineeringPrompt engineering is writing and testing the instructions and examples you give a model so its answer meets a goal you can check.4 min
2. System promptA system prompt is high-priority context that sets an AI assistant's role, boundaries, style, and response rules before the user's request is handled.3 min
3. Few-shot promptingFew-shot prompting places a small set of example inputs and outputs in the prompt so the model can imitate the task on a new input.3 min
4. Chain of thoughtChain-of-thought prompting asks for or demonstrates intermediate reasoning text before a final answer, but that text is not a guaranteed view of hidden model internals.4 min
5. Lost in the middleLanguage models often use facts near the opening or closing of a long input better than facts buried in the middle.5 min
6. Data privacyData privacy is about what an AI company may do with your chats, such as training models, and the settings that let you limit it.5 min

Let AI answer from your documents, not just memory.

1. EmbeddingsAn embedding maps a discrete item such as a token ID to a dense vector of numbers.3 min
2. Semantic searchSemantic search finds text by meaning, so a passage can match a question even when the two share no words.3 min
3. Vector databasesA vector database stores vectors and returns the saved records closest to a query vector, often with a score and any extra fields you ask for.4 min
4. RAGRAG lets an AI model look up relevant documents first, then answer from what it found.4 min

See how AI uses tools and works through tasks on its own.

1. AI agentsAn AI agent is a language model that chooses its own next steps and tools while it works through a task.4 min
2. Tool useTool use is a loop where a model proposes a call, a program outside the model runs it, and the result is written back.5 min
3. Function callingFunction calling lets a model request a named function by returning structured arguments that an application, or the provider, then runs.4 min
4. MCPMCP standardizes how AI applications connect to external context and tools.4 min
5. Agent stateAgent state is the running record an AI agent keeps while it works, so a task can pause, resume, or recover from a crash.5 min
6. Agent memoryAgent memory is how an AI agent saves useful information outside the model and loads the right pieces back into its prompt later.5 min
7. GuardrailsGuardrails are checks placed around an AI model that inspect what goes in and what comes out, and can block, change or check inputs and replies that are unsafe, off-topic or break the app's rules.5 min
8. Human in the loopWith this setup, a person checks an AI system's work at chosen points and can approve it, change it, reject it or step in.5 min
9. Prompt injectionPrompt injection is text that sneaks new instructions into what an AI model reads, so the model behaves in unintended ways.5 min

Know what LangChain, LangGraph, Google ADK and the agent SDKs each do, so you can pick one instead of writing everything yourself.

1. LangChainLangChain is an open-source framework for building apps and agents on top of large language models, with one standard way to talk to many AI providers.5 min
2. LlamaIndexLlamaIndex is an open-source toolkit that brings your own files and data to a language model when you ask a question, so apps and agents can answer from them.4 min
3. LangGraphLangGraph is an open-source framework for building AI agents as graphs, where steps share one state that can be saved, paused and resumed.5 min
4. Google ADKGoogle ADK is an open-source toolkit from Google for writing AI agents in code, giving them tools, a record of each chat and helper agents.5 min
5. OpenAI Agents SDKOpenAI's Agents SDK is a small open-source toolkit from OpenAI for building AI agents that use tools, pass work to each other and can record each run as a trace.5 min
6. Claude Agent SDKAnthropic's Claude Agent SDK is a Python and TypeScript library for building agents that run on the same loop, tools and context handling as Claude Code.5 min
7. CrewAICrewAI is an open-source toolkit, written in Python, for building teams of AI agents, each with a role, that work through a list of tasks together.4 min
8. Strands AgentsStrands Agents is a free AWS toolkit, released as open source under Apache 2.0, for making AI agents with Python or TypeScript, where the model itself plans the steps and picks the tools.5 min
9. Pydantic AIPydantic AI is a Python framework for building AI agents whose tool inputs and final answers are checked against types you define.4 min
10. Vercel AI SDKThe Vercel AI SDK is an open-source TypeScript toolkit, free to use, that lets apps talk to AI models from many companies through the same code.4 min

See how AWS, Google Cloud and Microsoft let you use, host and run models and agents without managing servers.

1. Amazon BedrockAmazon Bedrock is an AWS service that lets developers call AI models from several companies through one set of APIs, with extra tools for search and safety.5 min
2. Amazon Bedrock AgentCoreA set of AWS services, called Amazon Bedrock AgentCore, for running AI agents in the cloud, with hosting, memory, sign-in, tool access and monitoring.5 min
3. Amazon SageMaker AISageMaker AI, from AWS, lets you build, train and run machine learning models without managing your own servers.4 min
4. Vertex AIVertex AI is Google Cloud’s platform for choosing AI models, tuning them on your own examples and running models and agents for real users.5 min
5. Microsoft FoundryMicrosoft Foundry is Microsoft's Azure platform for building AI apps and agents, with a large model catalogue, tools, testing and safety controls in one place.5 min
6. Azure OpenAIAzure OpenAI lets companies use OpenAI's models through Microsoft's Azure cloud, with Azure billing, safety filters and data controls.5 min

Adapt, shrink, serve and monitor models once real people use them.

1. Inference optimizationInference optimization is a set of methods, like caching, quantization and speculative decoding, that let a trained language model answer with less work or less memory.3 min
2. Batch inferenceBatch inference sends many AI requests as one job that runs in the background, trading an instant reply for a lower price.3 min
3. Model servingModel serving means running a trained model on a server so apps can send it requests over the network and get answers back.5 min
4. LLM gatewaysAn LLM gateway is one server that sits between your apps and AI model providers, adding limits, caching, fallbacks and logs to model calls.5 min
5. Model routingModel routing sends each prompt to the model that suits it, so simple requests go to cheaper models and hard ones to stronger models.4 min
6. Fine-tuningFine-tuning takes a model that is already trained and keeps training it on a smaller set of examples for one task.4 min
7. QuantizationQuantization stores a model's numbers with fewer bits, such as 8-bit integers instead of 32-bit decimals, so it needs less memory.4 min
8. MLOpsMLOps is a way of working that makes building and releasing machine learning models simpler and more automatic.5 min
9. LLMOpsThe day-to-day work of running an app built on a language model, from managing prompts and testing answers to watching speed and cost.5 min

Compare a single agent loop, fixed workflows, graphs and teams of agents, and when each fits.

1. ReActReAct is a prompting method in which a language model alternates written thoughts with tool actions, reading each result before it picks the next step.5 min
2. Agentic workflowsAn agentic workflow runs a task as several language model and tool steps, joined by a path that your code fixes in advance.4 min
3. Agent planningAgent planning is how an AI agent splits a big task into smaller steps first, then carries out those steps one by one.4 min
4. Multi-agent systemsA multi-agent system is a group of AI agents, each a language model using tools in its own loop, that work together on one job.5 min
5. Sub-agentsA sub-agent is a helper AI that a main agent sends off to do one side task in its own workspace, then report back a short result.3 min
6. Agent orchestrationAgent orchestration is how an app with several AI agents decides which agent works, in what order, and who picks what happens next.4 min